S

Aw

taioaoiaisw[r

aabrep

1d

Biochemical Engineering Journal 41 (2008) 198–201

Contents lists available at ScienceDirect

Biochemical Engineering Journal

journa l homepage: www.e lsev ier .com/ locate /be j

hort communication

naerobic digestion of food waste in a hybrid anaerobic solid–liquid systemith leachate recirculation in an acidogenic reactor

anyan

ate iid and betreve

ogeniacido8,614intenty of

Olena Stabnikova ∗, Xue-Yan Liu, Jing-Yuan WangSchool of Civil and Environmental Engineering, Nanyang Technological University, 50 N

a r t i c l e i n f o

Article history:Received 17 March 2008Received in revised form 6 May 2008Accepted 21 May 2008

Keywords:Anaerobic digestionFood wasteHybrid anaerobic solid–liquid (HASL)systemRecirculationAcidogenic reactor

a b s t r a c t

Recirculation of the leachof food waste in the hybracidogenic reactor provideand buffering capacity to pof nutrients in the methanin the leachate from the9450 mg/l in control and 1in the acidogenic reactorproduce the same quantiwithout recirculation.

1. Introduction

Anaerobic digestion of solid food waste is considered as a per-spective way for its disposal [1,2]. To increase process efficiency,

wo-phase anaerobic digestion can be used [3]. Liquefactionnd acidification of food waste in this process are performedn an acidogenic reactor to obtain leachate containing dissolvedrganic matter, which is treated in a methanogenic reactor, wherecetogenesis and methanogenesis are running. As hydrolysis ofrganic matter is a slowest step in an anaerobic process [4], itsntensification usually leads to faster anaerobic digestion [5]. Therere known different methods to facilitate food waste hydrolysisn an acidogenic reactor, such as mechanical reduction of particleize of digested material [6,7]; thermal pre-treatment of foodaste to transform the organic solid material to liquid hydrolyzate

8–10]; and inoculation with cellulolytic bacteria from cattleumen [11].

It is known from the practice of organic waste landfill man-gement that a landfill leachate recirculation enhances microbialctivity and shorten time required for waste conversion and sta-ilization [12,13]. Therefore, using hypothetic analogy, leachateecirculation in an acidogenic reactor may be the way to enhancextraction of organic matter from the treated food waste in two-hase anaerobic digestion. This assumption was tested in the hybrid

∗ Corresponding author. Tel.: +65 6790 4740; fax: +65 6792 1650.E-mail address: costab@ntu.edu.sg (O. Stabnikova).

369-703X/$ – see front matter © 2008 Elsevier B.V. All rights reserved.oi:10.1016/j.bej.2008.05.008

g Avenue, Singapore 639798, Singapore

n the acidogenic reactor was proposed to enhance anaerobic digestionaerobic solid–liquid (HASL) system. Recirculation of the leachate in theter conditions for extraction of organic matter from the treated food wastent excessive acidification in the acidogenic reactor. It ensured faster supplyc reactor in experiment. The highest dissolved COD and VFA concentrationsgenic reactor were reached for shorter time and were 16,670 mg/l andmg/l and 11,094 mg/l in experiment, respectively. Recycling of the leachatesified anaerobic digestion of food waste and diminished time needed tomethane by 40% in comparison with anaerobic digestion of food waste

© 2008 Elsevier B.V. All rights reserved.

anaerobic solid–liquid (HASL) system, which was developed tominimize the amount of food waste for disposal in Singapore[14,15].

The present paper is a part of the research on HASL-enhancingperformance [10,16,17]. The aim of the present research was tostudy the effect of leachate recirculation in the acidogenic reac-tor on food waste anaerobic digestion in the hybrid anaerobic

solid–liquid system.

2. Materials and methods

2.1. Food waste, anaerobic microbial sludge and microbialgranules

Individual food waste was collected from a canteen in the univer-sity. Waste was shredded into particles with average size of 6.0 mmin a Robot-Coupe Shredder (CL50 Ultra, Hobart, France). The arti-ficial composition of food wastes used in the experiment was asfollows (% of wet weight): vegetable roots, 50; orange peels, 20; rice,15 and noodles, 15. The content of total solids (TS) was 20.3 ± 0.8%and the content of volatile solids (VS) was 76.2 ± 1.3% of TS in themixed waste.

Anaerobic microbial sludge, used as inoculum for the acido-genic and methanogenic reactors, was collected from an anaerobicdigester of a local wastewater treatment plant. The pH of sludgewas 7.1; the concentrations of the total solids and volatile solids insludge were 3.8 ± 0.1 g/l and 2.7 ± 0.1 g/l, respectively. The microbialanaerobic granules, adapted to high concentration of volatile fatty

O. Stabnikova et al. / Biochemical Engineering Journal 41 (2008) 198–201 199

leacha

Fig. 1. Schematic diagram of the HASL system operated without (a) and with (b)reactor; 3, peristaltic pump; 4, wet gas meter.

acids (VFA), were used as additional inoculum for the methanogenicreactor [18].

2.2. Characteristics and operation of the HASL system

The lab-scale HASL system includes an acidogenic reactorto treat solid food waste and an upflow anaerobic sludge blan-ket (UASB) methanogenic reactor to treat liquid leachate fromthe acidogenic reactor (Fig. 1a). Part of the effluent from themethanogenic reactor is used for the dilution of the acid effluentfrom the acidogenic reactor to maintain optimal pH for methano-genesis, and the rest of the effluent from the methanogenic reactoris recycled into the acidogenic reactor to avoid addition of water forfood waste hydrolysis. An inner diameter of the acidogenic reactorwas 140 mm and a height was 500 mm. Working volume of thereactor was 5.4 l. The working volume of the methanogenic reactorwas 3.0 l with inner diameter of 90 mm and a height of 600 mm.The HASL systems were operated in a constant temperature roomat 35 ± 1 ◦C.

Two sources of inoculum were used for the methanogenic reac-tor: anaerobic microbial sludge and microbial anaerobic granules.To start up the methanogenic reactor, suspension of microbialanaerobic granules, 1 l, and anaerobic microbial sludge, 1 l, were put

into the reactor, which was fed with synthetic wastewater [18]. Con-centration of chemical oxygen demand (COD) in effluent (less than500 mg COD/l) and average methane content in biogas (exceeding70% v/v) indicated that the methanogenic reactor was in active stateand ready for connection with the acidogenic reactor.

Control and experiment were carried out simultaneously in twoidentical HASL systems operated for 2 weeks. Food waste, 200 g, 1 lof distilled water and 36 g of CaCO3 for pH buffering were placed ineach acidogenic reactor. Anaerobic microbial sludge, 1 l, was addedinto each reactor as an inoculum. Food waste, 200 g, was added aftereach 24 h to the control and experimental reactors during first 4days of anaerobic digestion. So, organic loading rate for acidogenicreactors was 5.73 g VS/day during first 4 days of operation. Totally,800 g of food waste with the contents of total solids of 130.4 g andvolatile solids of 114.7 g were added into each acidogenic reactorin control and experiment. The flow rate of leachate from acido-genic reactor in experiment with recirculation was in 21 timeshigher than in control. However, volume of leachate supplied intothe methanogenic reactor was stable for both system 1440 ml/day(Fig. 1). Excess of leachate was recycled into acidogenic reactor inexperiment to ensure better conditions for extraction of organic

te recirculation in the acidogenic reactor: 1, acidogenic reactor; 2, methanogenic

matter from the treated food waste and to improve anaerobic pro-cess in the acidogenic reactor.

No leachate recirculation was in control, but leachate from theacidogenic reactor was recycled onto the top of waste in the acido-genic reactor in experiment (Fig. 1b). The ratio between volumeof recycled leachate and volume of leachate supplied into themethanogenic reactor was 20:1.

2.3. Chemical analysis

The leachate from the acidogenic reactor and the effluent fromthe methanogenic reactor were collected daily for analyses. The pHvalue was measured using a pH meter (CORNING 145, Halstead,Essex, England). Total solids, volatile solids, and COD of the feed-stock and digested food waste were determined in the well-mixedsamples in triplicates by standard methods [19]. For the determi-nation of VFA, samples were filtrated through Whatman 0.2 �mnitrocellulose membrane filters and were then analyzed usinghigh performance liquid chromatography (HPLC) (PerkinElmer,Series 200, Norwalk, CT, USA). The HPLC was equipped with a220 mm × 4.6 mm polypore H column and a UV 210 nm detector.The mobile phase was 0.005N H2SO4 with a flow rate at 0.15 ml/min.

Gas production was monitored by a wet gas meter (Ritter TG

05, Bochum, Germany), while gas composition was analyzed by aHewlett Packard GC HP5890A (HACH, Avondale, PA, and USA) formethane, carbon dioxide, and nitrogen. The GC was equipped with athermal conductivity detector and a stainless-steel column packedwith Hayesep Q (80/100 mesh). The operational temperatures ofinjector, detector and column were kept at 100 ◦C, 200 ◦C and 50 ◦C,respectively. Helium was used as a carrier gas at a flow rate of40 ml/min. The pH, concentrations of dissolved COD and VFA, aswell as biogas production and methane content, were determineddaily.

All analytical determinations were performed at least in tripli-cate and mean values ± S.D.s are shown.

3. Results and discussion

Food waste was placed in the acidogenic reactors of controland experimental HASL systems without or with leachate recir-culation, respectively. The results of anaerobic digestion in theHASL systems are shown in Fig. 2 and Table 1. The pH was lowerin control in comparison with experiment. The lowest pH in theleachate from the acidogenic reactor was 4.6 on day 5 in control

200 O. Stabnikova et al. / Biochemical Engineering Journal 41 (2008) 198–201

tor (Rin con

Fig. 2. pH (a) and concentration of VFA (b) in the leachate from the acidogenic reacin the effluent from the methanogenic reactor (Rm) (c), production of methane (d)

and 5.1 on day 6 in experiment (Fig. 2a). The peaks of VFA and dis-solved COD concentrations were higher and were reached in shorter

time in experiment in comparison with control. The highest VFAconcentrations in the leachate from the acidogenic reactors were9450 mg/l on day 5 in control and 11,094 mg/l on day 3 in exper-iment (Fig. 2b). Notwithstanding that concentrations of VFA werehigher in leachate from acidogenic reactor in experiment, pH ofleachate were lower than in control (Fig. 2a and b). As shown earlier[10], pH values in leachate from acidogenic reactor of batch HASLsystem dropped rapidly to pH about 5.0, probably due to cells andvacuoles lysis, and after that slowly increased to neutral pH, prob-ably due to hydrolysis of proteins and buffering of pH by releasedaminoacids. So, higher pH values of leachate in experiment could behypothetically explained by faster hydrolysis of proteins and fastertransfer the products of their hydrolysis in leachate under leachaterecirculation.

The highest dissolved COD concentrations in the leachate fromthe acidogenic reactors were 16,670 mg/l on day 6 in control and18,614 mg/l on day 5 in experiment (Fig. 2c). The concentrations ofdissolved COD in the effluent from the methanogenic reactor werelower in experiment than that in control (Fig. 2c). The same positiveeffect of leachate recycling with the pH adjustment in the one-stage landfill-simulating reactor was shown by Vavilin et al. [20].

Table 1Influence of leachate recirculation in the acidogenic reactor on the parameters ofanaerobic digestion of food waste during first 9 days in the HASL system

Parameters Control Experiment

Parameters of the acidogenic reactorDissolved COD produced (g) 98.7 ± 7.6 104.6 ± 6.7Maximum dissolved COD concentration (g/l) 16.7 ± 0.8 18.6 ± 0.9VFA produced (g) 59.9 ± 2.6 71.3 ± 3.8Maximum VFA concentration (g/l) 9.5 ± 0.5 11.1 ± 0.5VFA/dissolved COD (%) 62.9 71.3Lowest pH 4.6 5.1

Parameters of the methanogenic reactorMethane production (l) 23.7 28.6Average methane content (%) 72.2 ± 7.0 76.3 ± 6.2Maximum rate of methane production (l/day) 3.9 4.6

a), concentration of dissolved COD in the leachate from the acidogenic reactor andtrol (C) and in experiment (E). Error bars represent the standard deviations.

Total VFA, produced in the acidogenic reactor, and the VFA/CODratio were similar in experiment on day 9 and in control on day 14(Table 1).

The same volume of methane was produced for 14 days in con-trol and for 9 days in experiment (Fig. 2d, Table 1). Maximum ratesof biogas production were 3.9 l/day in control and 4.6 l/day in exper-iment. Therefore, recirculation shortened time for the productionof same volume of biogas.

According to the results, increase of total COD, VFA, and methaneproduction during batch anaerobic digestion of food waste for 14days in experiment was higher by 11% than in control. However,anaerobic process in the acidogenic reactor was faster in exper-iment, especially during first 6 days of operation. VFA and CODconcentrations in leachate from the acidogenic reactor on day3 were higher in experiment than that in control, by 37% and33%, respectively. Thus, recirculation of the leachate in the acido-genic reactor provided better conditions for extraction of organicmatter from the treated food waste and for the pH buffering

thus preventing excessive acidification in the acidogenic reac-tor.

The same amount of methane was generated during 14 days incontrol and during 9 days in experiment. This gives the opportunityto diminish operation time of batch process by 40%. It was supposedalso that recirculation will be more essential for semi-continuousanaerobic digestion, when fresh food waste will be added periodi-cally into acidogenic reactor on the top of semi-digested food waste.In this case, recirculation will play additional role as a source ofhydrolyzing and acidogenic bacteria for the treatment of fresh foodwaste.

The effect of leachate recirculation in the acidogenic reactor onfood waste anaerobic digestion in the hybrid anaerobic solid–liquidsystem was comparable with thermal pre-treatment of food wasteat 150 ◦C for 1 h which helped to diminish operation time ofbatch process in HASL by 50% [10] and with freezing/thawing pre-treatment of food waste (food waste was frozen for 24 h at −20 ◦Cand then thawed for 12 h at 25 ◦C) which gave the opportunity todiminish operational time of anaerobic digestion of food waste inHASL by 42% [17] in comparison with the anaerobic digestion offresh food waste.

Engine

[

[11] S.K. Han, H.S. Shin, Enhanced acidogenic fermentation of food waste in a

O. Stabnikova et al. / Biochemical

4. Conclusions

Recirculation of the leachate in the acidogenic reactor was pro-posed to enhance food waste anaerobic digestion in the hybridanaerobic solid–liquid system. Recirculation of the leachate in theacidogenic reactor provided better conditions for extraction oforganic matter from the treated food waste and enhanced bufferingcapacity preventing excessive acidification in the acidogenic reac-tor. It ensured faster supplying of nutrients in the methanogenicreactor in experiment. Use of leachate recirculation in the acido-genic reactor diminished time needed to produce the same quantityof methane by 40% in comparison with anaerobic digestion of foodwaste without recirculation. It was supposed that recirculation will

be more essential for semi-continuous anaerobic digestion thanfor batch process. Application of leachate recirculation in the aci-dogenic reactor of a two-phase anaerobic digestion system led tointensification of anaerobic digestion of food waste and did notdemand significant additional expenses.

Acknowledgments

The research was supported by a grant from the National Envi-ronment Agency of the Ministry of the Environment and WaterResources of the Republic of Singapore.

References

[1] W. Verstraete, D. de Beer, M. Pena, G. Lettinga, P. Lens, Anaerobic bioprocessingof organic wastes, World J. Microbiol. Biotechnol. 12 (1996) 221–238.

[2] M. Fehr, M.D.R. Calcado, D.C. Romao, The basis of a policy for minimizing andrecycling food waste, Environ. Sci. Policy 5 (2002) 247–253.

[3] F.G. Pohland, S. Ghosh, Developments in anaerobic treatments process, Biotech-nol. Bioeng. Symp. 2 (1971) 85–106.

[4] H.S. Shin, S.K. Han, Y.C. Song, C.Y. Lee, Multi-step sequential batch two-phaseanaerobic composting of food waste, Environ. Technol. 22 (2001) 271–279.

[

[

[

[

[

[

[

[

ering Journal 41 (2008) 198–201 201

[5] J.P. Strydom, T.J. Britz, J.F. Mostert, Two-phase anaerobic digestion of threedifferent dairy effluents using a hybrid bioreactor, Water SA 23 (1997) 151–156.

[6] S.K. Sharma, I.M. Mishra, M.P. Sharma, J.S. Saini, Effect of particle size on biogasgeneration from biomass residues, Biomass 17 (1988) 251–263.

[7] J. Mata-Alvarez, S. Mace, P. Llabres, Anaerobic digestion of organic solid wastes.An overview of research achievements and perspectives, Bioresour. Technol. 74(2000) 3–16.

[8] K. Tsukahara, T. Yagishita, T. Ogi, S. Sawayama, Treatment of liquid fractionseparated from liquidized food waste in an upflow anaerobic sludge blanketreactor, J. Biosci. Bioeng. 87 (1999) 554–556.

[9] D. Schieder, R. Schneider, F. Bischof, Thermal hydrolysis (TDH) as a pretreat-ment method for the digestion of organic waste, Water Sci. Technol. 41 (2000)181–187.

10] J.Y. Wang, X.Y. Liu, C.M. Kao, O. Stabnikova, Digestion of pre-treated food wastein hybrid anaerobic solid–liquid (HASL) system, J. Chem. Technol. Biotechnol.81 (2006) 345–351.

continuous-flow reactor, Waste Manage. Res. 20 (2002) 110–118.12] D.R. Reinhart, A.B. Al-Yousfi, The impact of leachate recirculation on municipal

solid waste landfill operating characteristics, Waste Manage. Res. 14 (1996)337–346.

13] F.G. Pohland, J.C. Kim, In situ anaerobic treatment of leachate in landfill biore-actors, Water Sci. Technol. 8 (1999) 203–220.

14] J.Y. Wang, H. Zhang, O. Stabnikova, S.S. Ang, J.H. Tay, A hybrid anaerobicsolid–liquid (HASL) system for food waste digestion, Water Sci. Technol. 52(2005) 223–228.

15] J.Y. Wang, H. Zhang, O. Stabnikova, J.H. Tay, Comparison of lab-scale andpilot-scale hybrid anaerobic solid–liquid systems operated in batch and semi-continuous modes, Process Biochem. 40 (2005) 3580–3586.

16] O. Stabnikova, S.S. Ang, X.Y. Liu, V. Ivanov, J.H. Tay, J.Y. Wang, The use of hybridanaerobic solid–liquid (HASL) system for the treatment of lipid-containing foodwaste, J. Chem. Technol. Biotechnol. 80 (2005) 455–461.

[17] O. Stabnikova, X.Y. Liu, J.Y. Wang, Digestion of frozen/thawed food waste in ahybrid anaerobic solid–liquid system, Waste Manage. (2007) (Published onlineSeptember 7, 2007).

18] J.Y. Wang, H. Zhang, O. Stabnikova, J.H. Tay, Removal of ammonia in a modifiedtwo-phase food waste anaerobic digestion system coupled with an aeratedsubmerged biofilters, Waste Manage. Res. 21 (2003) 527–534.

19] Standard Methods for the Examination of Water and Wastewater, 20th ed.,American Public Health Association/Water Environment Federation, Washing-ton, DC, 1998.

20] V.A. Vavilin, S.V. Rytov, L.Y. Lokshina, S.G. Pavlostathis, M.A. Barlaz, Distributedmodel of solid waste anaerobic digestion: effects of leachate recirculation andpH adjustment, Biotechnol. Bioeng. 81 (2003) 66–73.